• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.175-197
• /
• 2021

Recently, with the development of deep learning technology, research on unstructured data analysis is being actively conducted, and it is showing remarkable results in various fields such as classification, summary, and generation. Among various text analysis fields, text classification is the most widely used technology in academia and industry. Text classification includes binary class classification with one label among two classes, multi-class classification with one label among several classes, and multi-label classification with multiple labels among several classes. In particular, multi-label classification requires a different training method from binary class classification and multi-class classification because of the characteristic of having multiple labels. In addition, since the number of labels to be predicted increases as the number of labels and classes increases, there is a limitation in that performance improvement is difficult due to an increase in prediction difficulty. To overcome these limitations, (i) compressing the initially given high-dimensional label space into a low-dimensional latent label space, (ii) after performing training to predict the compressed label, (iii) restoring the predicted label to the high-dimensional original label space, research on label embedding is being actively conducted. Typical label embedding techniques include Principal Label Space Transformation (PLST), Multi-Label Classification via Boolean Matrix Decomposition (MLC-BMaD), and Bayesian Multi-Label Compressed Sensing (BML-CS). However, since these techniques consider only the linear relationship between labels or compress the labels by random transformation, it is difficult to understand the non-linear relationship between labels, so there is a limitation in that it is not possible to create a latent label space sufficiently containing the information of the original label. Recently, there have been increasing attempts to improve performance by applying deep learning technology to label embedding. Label embedding using an autoencoder, a deep learning model that is effective for data compression and restoration, is representative. However, the traditional autoencoder-based label embedding has a limitation in that a large amount of information loss occurs when compressing a high-dimensional label space having a myriad of classes into a low-dimensional latent label space. This can be found in the gradient loss problem that occurs in the backpropagation process of learning. To solve this problem, skip connection was devised, and by adding the input of the layer to the output to prevent gradient loss during backpropagation, efficient learning is possible even when the layer is deep. Skip connection is mainly used for image feature extraction in convolutional neural networks, but studies using skip connection in autoencoder or label embedding process are still lacking. Therefore, in this study, we propose an autoencoder-based label embedding methodology in which skip connections are added to each of the encoder and decoder to form a low-dimensional latent label space that reflects the information of the high-dimensional label space well. In addition, the proposed methodology was applied to actual paper keywords to derive the high-dimensional keyword label space and the low-dimensional latent label space. Using this, we conducted an experiment to predict the compressed keyword vector existing in the latent label space from the paper abstract and to evaluate the multi-label classification by restoring the predicted keyword vector back to the original label space. As a result, the accuracy, precision, recall, and F1 score used as performance indicators showed far superior performance in multi-label classification based on the proposed methodology compared to traditional multi-label classification methods. This can be seen that the low-dimensional latent label space derived through the proposed methodology well reflected the information of the high-dimensional label space, which ultimately led to the improvement of the performance of the multi-label classification itself. In addition, the utility of the proposed methodology was identified by comparing the performance of the proposed methodology according to the domain characteristics and the number of dimensions of the latent label space.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.51 no.2
• /
• pp.12-27
• /
• 2023

In this study, among the construction-type parks in Gangdong-gu, targeting parks with high environmental and ecological value located on the urban green axis, a plan was prepared for the ecological renewal of urban parks, and a design that applied to them was proposed. The renewal target site was selected by analyzing the general condition of Gangdong-gu and urban parks, the land use and green area ratio, park green area, and the green axis of Gangdong-gu. Gangdong-gu has 54 parks, including 2 neighborhood parks and 52 children's parks. In the first stage of the current status review, 17 parks were extracted through locational value analysis, such as location and adjacency to the natural axis and green axis. In the second stage, eight parks were selected among the first-stage extraction parks based on the ratio of green spaces and open spaces within each park service area. In the third stage, two of the second stage extraction parks were selected based on whether the legal standard of the park area was met, and in the fourth stage, one of the third stage extraction parks was selected through an aging survey of the park. As for the urban ecological status of the renewal target site, the status of land use in the aspect of entropy reduction, the status of soil cover in the aspect of water circulation, and the status of planting structure in the aspect of biodiversity were investigated. As for the status of the three renewal sites, the green area was insufficient at 18.3-45.3%, and the facility area was 54.7%-81.7%, which was judged to have low urban temperature reduction effects. The impervious pavement area accounted for 34.5% to 48.9% of the park area, accounting for most of the facility area, and it was judged that the water circulation function was insufficient. The planting structure consisted of a single layer and a double layer structure, and although the tree layer was good, the lower vegetation was poor, and there was no planting site of edible plants or large hardwood trees, so the biodiversity was low. After the ecological renewal design of Seonrin Children's Park, Dangmal Children's Park, and Saemmul Children's Park, which were selected as the renewal targets in this study, the ecological area ratio of each park increased by 1.4 to 3 times than before the renewal. If the urban parks located on the urban green axis are examined from the perspective of the urban ecosystem and renewed ecologically, it is judged that the expected effect will be high in reducing entropy, improving water circulation, and laying the foundation for biodiversity in terms of the urban ecosystem.

• Korean Journal of Heritage: History & Science
• /
• v.55 no.4
• /
• pp.180-193
• /
• 2022

The East Nine Royal Tombs is a representative place in the Royal Tombs of Joseon (a World Heritage Site). It consists of 1,289 stone artifacts including 979 related stone structures, 310 stone statues, and objects. Most of the stone structures in the East Nine Royal Tombs are composed of biotite granite, but some tombs are composed of light red granite. As a result of magnetic susceptibility measurement, the average data from Geonwolleung to Mongneung, excluding Hyeolleung, were similar, so it is estimated that stones were obtained from the same quarry. In the case of Sungneung, Sureung, and Gyeongneung, the range of susceptibility measurement is widely distributed. It assumed that the newly produced stones were mixed in the moving and construction process. Also, stones might be gathered from different quarries. As a result of a conservation status investigation, both the mound member and the ridge stone had the highest damage rate due to peeling and granular decomposition according to surface weathering. In the case of surface discoloration, yellowing and soils were found in the burial mound members. Yellowing, blackening, and soil were identified in the ridge stone structures. Bio-degradation is the major factor of deterioration of the East Nine Royal Tombs and the conservation status of the tombs were detected as grades 4 to 5. It seems that it is easy for the environment of the royal tombs to form soil for the microorganisms and fine conditions for continuous moisture. In the case of structures, they are in relatively good condition. As a result of a comprehensive damage rating for each tomb, the overall condition is good, but the Geonwolleung Royal Tomb and Hyeolleung Tomb, which were created in the early period, had relatively high weathering ratings. Stone objects in East Nine Royal Tombs have lost many pieces and gateway members due to surface deterioration. Also, secondary damage is ongoing. Each damage factor of the stone artifacts of the East Nine Royal Tombs combines to cause various and continuous damages. Therefore, it is necessary to establish regular conservation status data of the stone artifacts for efficient management after processing as well as conservation treatment of the royal tombs, and specific management manuals and systems. This study investigated the conservation status of stone structures in the East Nine Royal Tombs, a World Heritage Site, and systematically classified them to provide priority and necessity for conservation processing. We look forward to establishing a plan for the conservation and management of the East Nine Royal Tombs with this database in the future.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.20 no.1
• /
• pp.32-36
• /
• 2016

Purpose In nuclear medicine examination, $^{131}I$ is widely used in nuclear medicine examination such as diagnosis, treatment, and others of thyroid cancer and other diseases. $^{131}I$ conducts examination and treatment through emission of ${\gamma}$ ray and ${\beta}^-$ ray. Since $^{131}I$ (364 keV) contains more energy compared to $^{99m}Tc$ (140 keV) although it displays high integrated rate and enables quick discharge through kidney, the objective of this study lies in comparing the difference in exposure dose of $^{131}I$ before and after wearing apron when handling $^{131}I$ with focus on 3 elements of external exposure protection that are distance, time, and shield in order to reduce the exposure to technicians in comparison with $^{99m}Tc$ during the handling and administration process. When wearing apron (in general, Pb 0.5 mm), $^{99m}Tc$ presents shield of over 90% but shielding effect of $^{131}I$ is relatively low as it is of high energy and there may be even more exposure due to influence of scattered ray (secondary) and bremsstrahlung in case of high dose. However, there is no special report or guideline for low dose (74 MBq) high energy thus quantitative analysis on exposure dose of technicians will be conducted based on apron wearing during the handling of $^{131}I$. Materials and Methods With patients who visited Department of Nuclear Medicine of our hospital for low dose $^{131}I$ administration for thyroid cancer and diagnosis for 7 months from Jun 2014 to Dec 2014 as its subject, total 6 pieces of TLD was attached to interior and exterior of apron placed on thyroid, chest, and testicle from preparation to administration. Then, radiation exposure dose from $^{131}I$ examination to administration was measured. Total procedure time was set as within 5 min per person including 3 min of explanation, 1 min of distribution, and 1 min of administration. In regards to TLD location selection, chest at which exposure dose is generally measured and thyroid and testicle with high sensitivity were selected. For preparation, 74 MBq of $^{131}I$ shall be distributed with the use of $2m{\ell}$ syringe and then it shall be distributed after making it into dose of $2m{\ell}$ though dilution with normal saline. When distributing $^{131}I$ and administering it to the patient, $100m{\ell}$ of water shall be put into a cup, distributed $^{131}I$ shall be diluted, and then oral administration to patients shall be conducted with the distance of 1m from the patient. The process of withdrawing $2m{\ell}$ syringe and cup used for oral administration was conducted while wearing apron and TLD. Apron and TLD were stored at storage room without influence of radiation exposure and the exposure dose was measured with request to Seoul Radiology Services. Results With the result of monthly accumulated exposure dose of TLD worn inside and outside of apron placed on thyroid, chest, and testicle during low dose $^{131}I$ examination during the research period divided by number of people, statistics processing was conducted with Wilcoxon Signed Rank Test using SPSS Version. 12.0K. As a result, it was revealed that there was no significant difference since all of thyroid (p = 0.345), chest (p = 0.686), and testicle (p = 0.715) were presented to be p > 0.05. Also, when converting the change in total exposure dose during research period into percentage, it was revealed to be -23.5%, -8.3%, and 19.0% for thyroid, chest, and testicle respectively. Conclusion As a result of conducting Wilcoxon Signed Rank Test, it was revealed that there is no statistically significant difference (p > 0.05). Also, in case of calculating shielding rate with accumulate exposure dose during 7 months, it was revealed that there is irregular change in exposure dose for inside and outside of apron. Although the degree of change seems to be high when it is expressed in percentage, it cannot be considered a big change since the unit of accumulated exposure dose is in decimal points. Therefore, regardless of wearing apron during high energy low dose $^{131}I$ administration, placing certain distance and terminating the administration as soon as possible would be of great assistance in reducing the exposure dose. Although this study restricted $^{131}I$ administration time to be within 5 min per person and distance for oral administration to be 1m, there was a shortcoming to acquire accurate result as there was insufficient number of N for statistics and it could be processed only through non-parametric method. Also, exposure dose per person during lose dose $^{131}I$ administration was measured with accumulated exposure dose using TLD rather than through direct-reading exposure dose thus more accurate result could be acquired when measurement is conducted using electronic dosimeter and pocket dosimeter.

• Journal of Yeungnam Medical Science
• /
• v.8 no.1
• /
• pp.107-126
• /
• 1991

The purpose of the study was to assess the morbidity pattern and the medical care utilization behavior of the urban residents in the poor area. The study population included 2,591 family members of 677 households in the poor area of Daemyong 8 Dong, Nam-Gu, Taegu and 2,686 family members of 688 households, near the poor area in the same Dong, were interviewed as a control group. On this study the household interview method was applied. Well-trained interviewers visited every household in the designated area and individually interviewed heads of households or housewives for general information, morbidity condition, and medical care utilization with a structured questionnaire. Individuals were interviewed from 1 to 30 December 1988. The major results were summarized as follows : The proportion of the people below 5 years of age was 4.2% of the total study population and 5.5% were above 65 years of age in the poor area. This was slightly higher than in the control area. The average monthly income of a household in the poor area was 403,000 won versus 529,000 won in the control area. Fifty-eight percent of the residents in the poor area and sixty-one percent in the control area were medical security beneficiaries, but the proportion of medical aid beneficiaries was 7.8% in the poor area and 4.6% in the control area. The 15-day period morbidity rate of acute illnesses was 57.1 per 1,000 in the poor area and 24.2 per 1,000 in the control area. Respiratory disease is the most common acute illness in both areas. The most frequently utilized medical facility was the pharmacy among the patients with acute illnesses in the poor area. Among them 58.1% visited pharmacy initially while 38.4% of the patients in the control area visited a clinic. Among persons with illnesses during the 15 days 8.8% in the poor area and 4.6% in the control area did not seek any medical facility. Mean duration of utilization of medical facilities was 3.5 days in the poor area and 3.3 days in the control area. Initially of the medical facilities in Daemyong 8 Dong, The pharmacy in the poor area and the clinic in the control area were most commonly utilized. The most common reason for visiting the hospital was 'regular customers' in the poor area and 'geographical accessibility' in the control area. The one year period morbidity rate of chronic illness in the poor area was 83.0 per 1,000 population and 28.0 per 1,000 in the control area. Disease of nervous system was the most common chronic illness in the poor area while cardiovascular disease in male and gastrointestinal disease in female were most prevalent in the control area. The most frequently utilized medical facility was the pharmacy among the patients with chronic illnesses in the poor area. Among them 24.2% visited the pharmacy initially while 34.7% of the patients in the control area visited the out-patient department of the hospital within a 15-day period. Among the patients with chronic illnesses 34.9% in the poor area and 16.0% in the control area did not seek any medical facility. Mean duration of utilization of medical facilities was 9.2 days in the poor area and 9.9 days in the control area within a 15-day period. Initially of the medical facilities in Daemyong 8 Dong, the pharmacy in the poor area and the hospital in the control area were most commonly utilized. The most common reason for visiting the hospital, clinic, health center or pharmacy in the poor area was 'geographical accessibility' while the reason for visiting herb clinic was 'good result' and 'reputation' in both areas.

• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.1-33
• /
• 2019

The small and medium sized enterprises (hereinafter SMEs) are already at a competitive disadvantaged when compared to large companies with more abundant resources. Manufacturing SMEs not only need a lot of information needed for new product development for sustainable growth and survival, but also seek networking to overcome the limitations of resources, but they are faced with limitations due to their size limitations. In a new era in which connectivity increases the complexity and uncertainty of the business environment, SMEs are increasingly urged to find information and solve networking problems. In order to solve these problems, the government funded research institutes plays an important role and duty to solve the information asymmetry problem of SMEs. The purpose of this study is to identify the differentiating characteristics of SMEs that utilize the public information support infrastructure provided by SMEs to enhance the innovation capacity of SMEs, and how they contribute to corporate performance. We argue that we need an infrastructure for providing information support to SMEs as part of this effort to strengthen of the role of government funded institutions; in this study, we specifically identify the target of such a policy and furthermore empirically demonstrate the effects of such policy-based efforts. Our goal is to help establish the strategies for building the information supporting infrastructure. To achieve this purpose, we first classified the characteristics of SMEs that have been found to utilize the information supporting infrastructure provided by government funded institutions. This allows us to verify whether selection bias appears in the analyzed group, which helps us clarify the interpretative limits of our study results. Next, we performed mediator and moderator effect analysis for multiple variables to analyze the process through which the use of information supporting infrastructure led to an improvement in external networking capabilities and resulted in enhancing product competitiveness. This analysis helps identify the key factors we should focus on when offering indirect support to SMEs through the information supporting infrastructure, which in turn helps us more efficiently manage research related to SME supporting policies implemented by government funded institutions. The results of this study showed the following. First, SMEs that used the information supporting infrastructure were found to have a significant difference in size in comparison to domestic R&D SMEs, but on the other hand, there was no significant difference in the cluster analysis that considered various variables. Based on these findings, we confirmed that SMEs that use the information supporting infrastructure are superior in size, and had a relatively higher distribution of companies that transact to a greater degree with large companies, when compared to the SMEs composing the general group of SMEs. Also, we found that companies that already receive support from the information infrastructure have a high concentration of companies that need collaboration with government funded institution. Secondly, among the SMEs that use the information supporting infrastructure, we found that increasing external networking capabilities contributed to enhancing product competitiveness, and while this was no the effect of direct assistance, we also found that indirect contributions were made by increasing the open marketing capabilities: in other words, this was the result of an indirect-only mediator effect. Also, the number of times the company received additional support in this process through mentoring related to information utilization was found to have a mediated moderator effect on improving external networking capabilities and in turn strengthening product competitiveness. The results of this study provide several insights that will help establish policies. KISTI's information support infrastructure may lead to the conclusion that marketing is already well underway, but it intentionally supports groups that enable to achieve good performance. As a result, the government should provide clear priorities whether to support the companies in the underdevelopment or to aid better performance. Through our research, we have identified how public information infrastructure contributes to product competitiveness. Here, we can draw some policy implications. First, the public information support infrastructure should have the capability to enhance the ability to interact with or to find the expert that provides required information. Second, if the utilization of public information support (online) infrastructure is effective, it is not necessary to continuously provide informational mentoring, which is a parallel offline support. Rather, offline support such as mentoring should be used as an appropriate device for abnormal symptom monitoring. Third, it is required that SMEs should improve their ability to utilize, because the effect of enhancing networking capacity through public information support infrastructure and enhancing product competitiveness through such infrastructure appears in most types of companies rather than in specific SMEs.

• Proceedings of the Safety Management and Science Conference
• /
• 2004.11a
• /
• pp.355-380
• /
• 2004

I test the hypothesis that the gradual diffusion of information across asset markets leads to cross-asset return predictability in Korea. Using thirty-six industry portfolios and the broad market index as our test assets, I establish several key results. First, a number of industries such as semiconductor, electronics, metal, and petroleum lead the stock market by up to one month. In contrast, the market, which is widely followed, only leads a few industries. Importantly, an industry's ability to lead the market is correlated with its propensity to forecast various indicators of economic activity such as industrial production growth. Consistent with our hypothesis, these findings indicate that the market reacts with a delay to information in industry returns about its fundamentals because information diffuses only gradually across asset markets. Traditional theories of asset pricing assume that investors have unlimited information-processing capacity. However, this assumption does not hold for many traders, even the most sophisticated ones. Many economists recognize that investors are better characterized as being only boundedly rational(see Shiller(2000), Sims(2201)). Even from casual observation, few traders can pay attention to all sources of information much less understand their impact on the prices of assets that they trade. Indeed, a large literature in psychology documents the extent to which even attention is a precious cognitive resource(see, eg., Kahneman(1973), Nisbett and Ross(1980), Fiske and Taylor(1991)). A number of papers have explored the implications of limited information- processing capacity for asset prices. I will review this literature in Section II. For instance, Merton(1987) develops a static model of multiple stocks in which investors only have information about a limited number of stocks and only trade those that they have information about. Related models of limited market participation include brennan(1975) and Allen and Gale(1994). As a result, stocks that are less recognized by investors have a smaller investor base(neglected stocks) and trade at a greater discount because of limited risk sharing. More recently, Hong and Stein(1999) develop a dynamic model of a single asset in which information gradually diffuses across the investment public and investors are unable to perform the rational expectations trick of extracting information from prices. Hong and Stein(1999). My hypothesis is that the gradual diffusion of information across asset markets leads to cross-asset return predictability. This hypothesis relies on two key assumptions. The first is that valuable information that originates in one asset reaches investors in other markets only with a lag, i.e. news travels slowly across markets. The second assumption is that because of limited information-processing capacity, many (though not necessarily all) investors may not pay attention or be able to extract the information from the asset prices of markets that they do not participate in. These two assumptions taken together leads to cross-asset return predictability. My hypothesis would appear to be a very plausible one for a few reasons. To begin with, as pointed out by Merton(1987) and the subsequent literature on segmented markets and limited market participation, few investors trade all assets. Put another way, limited participation is a pervasive feature of financial markets. Indeed, even among equity money managers, there is specialization along industries such as sector or market timing funds. Some reasons for this limited market participation include tax, regulatory or liquidity constraints. More plausibly, investors have to specialize because they have their hands full trying to understand the markets that they do participate in

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.1-20
• /
• 2013

Nowadays, social network is a huge communication platform for providing people to connect with one another and to bring users together to share common interests, experiences, and their daily activities. Users spend hours per day in maintaining personal information and interacting with other people via posting, commenting, messaging, games, social events, and applications. Due to the growth of user's distributed information in social network, there is a great potential to utilize the social data to enhance the quality of recommender system. There are some researches focusing on social network analysis that investigate how social network can be used in recommendation domain. Among these researches, we are interested in taking advantages of the interaction between a user and others in social network that can be determined and known as social relationship. Furthermore, mostly user's decisions before purchasing some products depend on suggestion of people who have either the same preferences or closer relationship. For this reason, we believe that user's relationship in social network can provide an effective way to increase the quality in prediction user's interests of recommender system. Therefore, social relationship between users encountered from social network is a common factor to improve the way of predicting user's preferences in the conventional approach. Recommender system is dramatically increasing in popularity and currently being used by many e-commerce sites such as Amazon.com, Last.fm, eBay.com, etc. Collaborative filtering (CF) method is one of the essential and powerful techniques in recommender system for suggesting the appropriate items to user by learning user's preferences. CF method focuses on user data and generates automatic prediction about user's interests by gathering information from users who share similar background and preferences. Specifically, the intension of CF method is to find users who have similar preferences and to suggest target user items that were mostly preferred by those nearest neighbor users. There are two basic units that need to be considered by CF method, the user and the item. Each user needs to provide his rating value on items i.e. movies, products, books, etc to indicate their interests on those items. In addition, CF uses the user-rating matrix to find a group of users who have similar rating with target user. Then, it predicts unknown rating value for items that target user has not rated. Currently, CF has been successfully implemented in both information filtering and e-commerce applications. However, it remains some important challenges such as cold start, data sparsity, and scalability reflected on quality and accuracy of prediction. In order to overcome these challenges, many researchers have proposed various kinds of CF method such as hybrid CF, trust-based CF, social network-based CF, etc. In the purpose of improving the recommendation performance and prediction accuracy of standard CF, in this paper we propose a method which integrates traditional CF technique with social relationship between users discovered from user's behavior in social network i.e. Facebook. We identify user's relationship from behavior of user such as posts and comments interacted with friends in Facebook. We believe that social relationship implicitly inferred from user's behavior can be likely applied to compensate the limitation of conventional approach. Therefore, we extract posts and comments of each user by using Facebook Graph API and calculate feature score among each term to obtain feature vector for computing similarity of user. Then, we combine the result with similarity value computed using traditional CF technique. Finally, our system provides a list of recommended items according to neighbor users who have the biggest total similarity value to the target user. In order to verify and evaluate our proposed method we have performed an experiment on data collected from our Movies Rating System. Prediction accuracy evaluation is conducted to demonstrate how much our algorithm gives the correctness of recommendation to user in terms of MAE. Then, the evaluation of performance is made to show the effectiveness of our method in terms of precision, recall, and F1-measure. Evaluation on coverage is also included in our experiment to see the ability of generating recommendation. The experimental results show that our proposed method outperform and more accurate in suggesting items to users with better performance. The effectiveness of user's behavior in social network particularly shows the significant improvement by up to 6% on recommendation accuracy. Moreover, experiment of recommendation performance shows that incorporating social relationship observed from user's behavior into CF is beneficial and useful to generate recommendation with 7% improvement of performance compared with benchmark methods. Finally, we confirm that interaction between users in social network is able to enhance the accuracy and give better recommendation in conventional approach.